• DocumentCode
    2552136
  • Title

    Efficient parameter quantisation for 2.4/1.2 kb/s split-band LPC coding

  • Author

    Villette, S. ; Cho, Y.D. ; Kondoz, A.M.

  • Author_Institution
    Centre for Commun. Syst. Res., Surrey Univ., Guildford, UK
  • fYear
    2000
  • fDate
    2000
  • Firstpage
    32
  • Lastpage
    34
  • Abstract
    Speech coding at very low bit rates has many applications such as answering machines, IP telephony, mobile communications, military communications etc. Most low bit rate coders operate at around 2.4 kb/s, as the speech quality degrades too much below this bit rate. We describe a frequency domain speech coder capable of operating at both 2.9 and 1.2 kb/s, and produces good quality synthesised speech. Both rates use the same analysis and synthesis building blocks over 20 ms, but the 1.2 kb/s coder jointly quantises three sets of parameters every 60 ms to reduce the bit rate while maintaining speech quality. We also describe the quantisation methods used to lower the bit rate from 2.4 kb/s to 1.2 kb/s while retaining most of the quality of the higher bit rate version
  • Keywords
    linear predictive coding; quantisation (signal); speech coding; speech intelligibility; speech synthesis; 1.2 kbit/s; 2.4 kbit/s; 2.9 kbit/s; 20 ms; 60 ms; IP telephony; answering machines; bit rate reduction; efficient parameter quantisation; frequency domain speech coder; low bit rate coders; military communications; mobile communications; quantisation methods; speech analysis; speech coding; speech quality; speech synthesis; split-band LPC coding; synthesised speech quality; very low bit rates; Bit rate; Degradation; Linear predictive coding; Military communication; Mobile communication; Quantization; Speech analysis; Speech coding; Speech synthesis; Telephony;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Speech Coding, 2000. Proceedings. 2000 IEEE Workshop on
  • Conference_Location
    Delavan, WI
  • Print_ISBN
    0-7803-6416-3
  • Type

    conf

  • DOI
    10.1109/SCFT.2000.878385
  • Filename
    878385